See

discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/245188387

Concurrent redesign of an underactuated robot

manipulator
Article in Mechatronics March 2009
DOI: 10.1016/j.mechatronics.2008.09.002

CITATIONS

READS

11

50

3 authors:
Carlos Alberto Cruz Villar

Jaime lvarez

Center for Research and Advanced Studies

Center for Research and Advanced Studies

51 PUBLICATIONS 182 CITATIONS

125 PUBLICATIONS 710 CITATIONS

SEE PROFILE

SEE PROFILE

Miguel G. Villarreal-Cervantes
Instituto Politcnico Nacional
25 PUBLICATIONS 66 CITATIONS
SEE PROFILE

All in-text references underlined in blue are linked to publications on ResearchGate,

letting you access and read them immediately.

Available from: Jaime lvarez

Retrieved on: 22 October 2016

Mechatronics 19 (2009) 178183

Contents lists available at ScienceDirect

Mechatronics
journal homepage: www.elsevier.com/locate/mechatronics

Concurrent redesign of an underactuated robot manipulator

Carlos A. Cruz-Villar, Jaime Alvarez-Gallegos, Miguel G. Villarreal-Cervantes *
Mechatronics Section, Electrical Engineering Department, Cinvestav-IPN, Av. Instituto Politcnico Nacional, No. 2508, Col. San Pedro Zacatenco, Mxico DF 07360,
Apdo. Postal 14-740, Mexico

a r t i c l e

i n f o

Article history:
Received 15 May 2007
Accepted 1 September 2008

Keywords:
Concurrent redesign
Design optimization
Minimum time
Pendubot

a b s t r a c t
An interesting problem in robotics is to minimize the required time to force a manipulator to travel
between two specic points (positioning time). In this paper a concurrent structure-control redesign
approach is proposed in order to nd the minimum positioning time of an underactuated robot manipulator, by considering a synergetic combination between the structural parameters and a bangbang control law. The problem consists in nding the structural parameters of the system and the switching
intervals of the bangbang control that simultaneously minimize the positioning time, subject to the
input constraint and the structural parameter constraint. The concurrent structure-control redesign
approach is stated as a dynamic optimization problem (DOP). The projected gradient method is used
to solve the DOP.
The effectiveness of the proposed concurrent redesign approach is shown via simulation and experimental results on an underactuated system called the Pendubot.
2008 Elsevier Ltd. All rights reserved.

1. Introduction
Robots are currently used for a variety of tasks in industrial
applications. Improving their performance could result in commercial benets.
Time-optimal control of robots [1,2] has been used in order to
minimize the required time to take the system from an initial state
to a nal one (positioning time). To nd the time-optimal control,
innite dimensional problems are usually stated. A variational approach, which focuses on obtaining a solution via the classical necessary conditions for optimality, solves the innite dimensional
problem. It is proven in [4] that the behavior of the control signal
for time-optimal control problems with bounded control input and
with terminal constraints is of bangbang nature.1 Then, timeoptimal control problems can be parametrized [3] in order to
transform them into nite dimensional problems. So these nite
dimensional problems can be solved by using nonlinear programming (NLP) methods.
On the other hand, in the last years, an integrated design approach [58] which involves a structure-control design, has been
employed to improve the performance of electromechanical systems. This approach consists in integrating the design parameters
* Corresponding author. Tel.: +52 5550613800x6311.
E-mail addresses: gvillarr@cinvestav.mx, villamike@hotmail.com (M.G. Villarreal-Cervantes).
URL: http://www.meca.cinvestav.mx/ (M.G. Villarreal-Cervantes).
1
A control signal which is restricted to be in a lower and an upper bound and it
switches immediately from one extreme to another at certain times (the control
signal is never strictly between the bounds), is called bangbang control.
0957-4158/$ - see front matter 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.mechatronics.2008.09.002

of both the mechanical structure and the controller into a single

design process.
Fu and Mills [5] proposed a structure-control design approach
based on convex optimization theory, a linear model of the system
and closed-loop control laws. In that approach, the design specications (performance index vector) must be convex with respect to
the closed-loop transfer function. Rastegar et al. [6] optimized the
kinematic, dynamic and closed-loop control parameters with an
integrated design approach in order to obtain a high performance
robotic system that rapidly executes point-to-point motions, with
minimal vibration. Ravichandran et al. [7] presented a methodology based on numerical optimization techniques for simultaneously optimizing design parameters of a two-link planar rigid
manipulator and nonlinear PD controllers. Both design parameters
are optimized to perform multiple tasks. In Alvarez et al. [8] a
structure-control integrated approach is used in order to design a
pinion-rack continuously variable transmission. A multi objective
dynamic optimization problem is stated in order to concurrently
obtain a set of optimal mechanical structure and control
parameters.
The previous works share a common characteristic: the parameters of a closed-loop control law were taken into account in the
integrated design approach. In this paper, a concurrent redesign
approach is proposed. This approach involves a parametric redesign procedure of the system considering structural and controller
parameters. Hence the open-loop bangbang control parameters
and the structural parameters are simultaneously found in this approach. Therefore, the concurrent redesign approach is formulated
as a dynamic optimization problem, where the design variables are

179

C.A. Cruz-Villar et al. / Mechatronics 19 (2009) 178183

both the structural parameters of the system and the switching

time interval vector of the bangbang control law and the redesign
objective is to minimize the positioning time of a nonlinear electromechanical system, subject to the system dynamics. The
numerical projected gradient optimization method is used to solve
such a dynamic optimization problem. To apply this approach, it is
assumed that the trajectory from the initial state to the nal one
must be obstacle free, that is, without trajectory restrictions and
that dynamic system model is a smooth and differentiable one.
Underactuated systems arise in many circumstances such as
when the system is underactuated by design, or when the actuators are expensive and/or heavy, such that they are avoided in
the system design, or due to an actuator failure. If an actuator fails,
a solution via hardware or software could be implemented. A solution via hardware may be performed by equipping the system with
redundant actuators, adding in consequence weight to the system.
On the other hand, a solution via software is a cost and weightreducing alternative, since only a switching of the control law is required when an actuator failure is detected. So, control laws for
underactuated manipulators could provide fault tolerance to the
system against actuator failures [9].
In this paper, it is proposed a concurrent redesign approach to
the swinging up problem of an underactuated double pendulum
system called Pendubot.
The remainder of this paper is organized as follows, the concurrent redesign approach is presented in Section 2. Dynamic equations of the Pendubot and link parameters (masses, mass center
lengths, inertias) with the structural modications are found in
Section 3. Simulation and experimental results are found in Section
4. Conclusions of this paper are presented in Section 5.
2. Concurrent redesign approach for system positioning in
minimum time
The concurrent redesign approach consists in nding the input
control signal ut 2 R for all time t0 6 t 6 tf and the structural
parameters  2 Rn of an electromechanical system, that bring
the system from the initial conditions xt 0 x0 to the nal one
xt f xf in minimum time. In this paper, the redesign problem
is formulated as a dynamic optimization problem (DOP) consisting
of minimizing under ut and  , the functional (1) subject to the
nonlinear dynamic equation of the electromechanical system (2),
the initial state (3), the nal state (4), the input control signal constraint (5) and the structural parameter constraints (6), where
xt 2 Rn is the state vector, ut is the input control signal,  is
the structural parameter vector, x0 and xf are the initial and the nal desired states, respectively

J tf
x_ f xt; ut; 

xt0 x0

xtf xf

umin 6 ut 6 umax

 min 6  6  max

To transform the constrained problem stated above into an unconstrained one, barrier functions [10] are considered. These barrier
functions set a barrier against the solution leaves the feasible region
imposed by the constraints when they are added to the performance index of the optimization problem. Therefore, the dynamic
optimization problem stated in (1)(6) is reformulated in (8)(11),
which considers the inequality constraints (6) as barrier functions.
Moreover, dening the input control vector ut as a bangbang
control law, the switching time interval vector Dt of the control signal must be included in the next problem reformulation. Then ut
is stated in the following way:

ut

umax

if Dt j > 0

umin

if Dt j < 0

The sign of the switching interval Dt j , obtained by the optimization

algorithm (12), could have a positive or negative sign representing
the corresponding control signal (umin or umax ), and the absolute value of the switching interval jDt j j represents the time interval length
when the corresponding control signal (umin or umax ) is applied.
Then, the problem reformulation of the dynamic optimization
problem (1)(6) consists in nding the switching time intervals
of the control signal Dt Dt 1 ; . . . ; DtnDt T 2 RnDt and the structural
parameters   1 ; . . . ;  n T 2 Rn that minimize the modied
performance index (8) subject to the nonlinear dynamic equation
of the electromechanical system (9), the initial conditions (10)
and the nal state xtf constraints (11), with ut as dened in
P Dt
 Pn Nl is the barrier function
jDtj j. In Eq. (8), N
(7) and tf  nj1
l1
set
for
the
structural
parameter
vector
,
where
h
i
1
1
Nl x l  

is
the
barrier
function
for
the
 

lmax

lmin

l  th structural parameter,  min  1min ; . . . ;  n min T 2 Rn and

 max  1max ; . . . ;  n max T 2 Rn are the minimum and maximum
limit
vectors
for
the
structural
parameter

and
x x 1 ; . . . ; x n  2 Rn is a constant vector with elements as
close to zero as possible, according to the barrier function approach
to deal with inequality constraints [10]


JDt;  t f N

nDt
n
X
  X
 Dt j 
Nl
j1

l1

x_ f xt; Dt; 

xt 0 x0

10

ef xt f  xf 0

11

As equation (8) and (9) depend explicitly on the parameters Dt and

 , changes in those parameters would modify the system
performance.
In this work, the projected gradient algorithm (12) is used to
solve the problem stated above, where kr ; kn 2 RnDt n nDt n are
diagonal matrices with the step sizes krii ; knii for i 1; :::; nDt n
in the diagonal elements. I is the identity matrix,
 oxtf ; oxtf  2 RnnDt n is the Jacobian matrix of partial derivaA
oD t
o
ot

tives of xt f with respect to Dt and  ; rJ oDft ; ooN T is the gradient


of J with respect to Dt and  ; P I  A A is the projection matrix
of the gradient of the performance index onto the null space of the
 A
 is the MoorePenrose pseudo-inverse of
constraint matrix A;

T   T 1
 1 A
T

 A
 T A


if
n 6 nDt n ,
or
A
if
AA A AA
 ef is the solution that minimizes the error
n > nDt n , and A
ef xtf  xf . The convergence of (12) to a solution is guaranteed
with small step sizes kr and kn [10], that is, the constraint error ef
and the projection P DJ converge to zero. Then, the step sizes krii ; knii
are selected to be in the interval (0, 1]. In order to obtain the optimal parameters with less computing time, the constraint error fulllment ef must have more importance than the projection P DJ in
(12), that is, krii P knii must be taken into account

  
Dt new
Dt
 ef  kn PrJ

 kr A
 new


12

The sensitivity of xt f with respect to each component Dt j (13) of

the switching interval vector Dt can be obtained as in [1], where
sgn is the sign function, xj  xt j and x_ j f xj t; Dt j ;  .
ox
oxt
Aj t j ox j , ox f AnDt AnDt 1 . . . Aj1 by the chain rule
j1

oxtf oxtf otj oxj oxtf

x_ j sgnDt j
o Dt j
ot j oDt j oxj
oxj

13

Hence, the matrices Aj can be computed by integrating the rst varoxt

; t j1 6 t 6 t j , then Aj satises (14), with
iation of (9). If Aj t ox
j1
the initial condition Aj t j1 I for each t j1 6 t 6 tj , where I is the

180

C.A. Cruz-Villar et al. / Mechatronics 19 (2009) 178183

identity matrix, Aj Aj tj and xt is the trajectory resulting from

Dt



dAj t
of xt; Dt j ; 

Aj t
ot
ox

14

The sensitivity of xtf with respect to each structural parameter  l

0
can be obtained from (15), with the initial condition oxt
0
o

 
d oxt
o
l

dt



of xt ; Dt;  oxt
of xt; Dt; 

oxt
o l
o l

15

In this work the mechanism proposed in [1] is used to add or delete

switching intervals in the projected gradient algorithm (12). As the
concurrent redesign optimization problem is a nonlinear one, the
projected gradient algorithm (12) with different starting conditions
Dt ini ;  ini could converge to different local minima Dt H ;  H .
Therefore, a procedure to select appropriate starting conditions
should be used.
3. Dynamic system and structural parameters
The Pendubot (Fig. 1) is an underactuated double pendulum
system with one actuator at the rst joint and no actuator at the
second one. The lack of an actuator severely modies the dynamic
behavior of the system which, may become chaotic [12].
The Pendubot proposal was in 1991 [13]. Since that time, one
control objective has been to bring the links from a given initial
conguration, namely the downwards position or rest position,
to an arbitrary nal position, in particular the upwards position.
The swinging up control consists in bringing the links from the
downwards position (q1 p=2 rad; q2 0 rad, in Fig. 1) to
the upwards position (q1 p=2rad; q2 0 rad, in Fig. 1). Another control law, which is called the balance control, is then required in order to stabilize the links in the upwards position. In
this paper the swinging up control is the bangbang control obtained via the concurrent redesign approach and the balance control is a linear quadratic regulator (LQR). Block [13] proposed a
strategy for switching from the swinging up control to the balance
control at the right moment. This switching strategy consists in
dening a neighborhood where the links are considered at the upwards position or in other words, in the basin of attraction of the
lineal controller (LQR). When the links are in the neighborhood,
the control law must be switched. In this paper, such a neighborhood is called the attraction region of the LQR.
The development of the dynamic equations of the Pendubot can
be found in [9], which can be written as in (16), where s s1 ; 0T is

Fig. 1. Schematic representation of the Pendubot.

q
1 ; q
2 T
the motor torque vector, q q1 ; q2 T ; q_ q_ 1 ; q_ 2 T and q
are the position, velocity and acceleration vectors of the links 1
and 2, respectively

_ q_ Gq
s Dqq Cq; q

16

Since in this work the nature of the open-loop control law is bang
bang and as the actuator (a DC armature controlled motor) can not
deliver a perfect bangbang signal, the dynamics of the motor (the
electrical constant time se and the mechanical constant time sm )
must be considered. Hence, the Pendubot dynamics (16) coupled
to the motor dynamics is taken into account and it can be found
in [14]. As it is stated in Section 2, a nonlinear state representation
of the system (2) is considered, where x q1 ; q2 ; q_ 1 ; q_ 2 , ia T is the
state vector, ia is the armature motor current and ut V in ; 0T is
the input voltage vector.
In Fig. 1, the schematic representation of the Pendubot is
shown, where m1 ; m2 ; l1 ; l2 ; lc1 , lc2 ; I1 ; I2 are the masses, lengths,
lengths between the rotation axis and the center of mass, and
the inertias of links 1 and 2, respectively.
To implement the concurrent redesign approach, it is necessary
to modify the structural parameters of the Pendubot. Then by considering that the Pendubot is already built, a weight of cylindrical
shape is placed at each link in order to modify the dynamic parameters mi ; lci ; Ii ; i 1; 2 of the Pendubot. These cylindrical weights
are shown in Fig. 2, where mp1 ; mp2 ; lcp1 ; lcp2 are the masses and
the distance between the rotation axis and the mass center of
cylindrical weights 1 and 2. Thus,  mp1 Kg; mp2
Kg; lcp1 m; lcp2 mT 2 R4 is the structural parameter vector to
be modied.
The dynamic parameters of the Pendubot with the added cylindrical weights are obtained via classical methods [15], and summarized in Table 1. Note that these dynamic parameters
~ i; f
lci ; ~Ii ; i 1; 2 are functions of both the constant dynamic
m
parameters of the Pendubot m1 0:8435 Kg; m2 0:3859
Kg;lc1 0:1573 m;lc2 0:1416 m;I1 0:0057 Kgm2 ;I2 0:00539
Kgm2  and the parameters of the cylindrical weights which are
e 1 2 Kgm2 ;
grouped in  , where Izp1 12 mp1 rp1 2 mp1 lcp1  lc
2
2
2
1
e
Kgm  and rp1 0:0254 m,
Izp2 mp2 rp2 mp2 lcp2  lc 2
2

rp2 0:0254 m are the inertias and the radii of cylindrical

weights on links 1 and 2, respectively.
It is important to note that changes in the structural parameters
of the cylindrical weights  modify the dynamic parameters of

Fig. 2. CAD drawing of the Pendubot with the added cylindrical weights.

181

C.A. Cruz-Villar et al. / Mechatronics 19 (2009) 178183

Table 1
Dynamic parameters of the Pendubot with the cylindrical weights
Link 1

Link 2

~ 1 0:8435 mp1 Kg

m
e 0:1573 lcp1 mp1 m
lc
1
~1
m
~I1 0:0057 Izp1 Kgm2 

~ 2 0:3859 mp2 Kg

m
e 0:1416 lcp2 mp2 m
lc
2
~2
m
~I2 0:00539 Izp2 Kgm2 

~ i; f
the whole system m
lci ; ~Ii ; i 1; 2. Now, the motion equation of
the Pendubot x_ f xt; Dt;  is a function of the state vector
xt, the switching time interval vector Dt of the control signal
ut and the structural parameter vector  of the cylindrical
weights.
Changes in mpi for i 1; 2 must be positive because mass can
only be added to the already constructed Pendubot and changes
in lcpi for i 1; 2 could be positive or negative. A negative change
in lcpi means that the displacement of the cylindrical weight is
on the x-axis (see Fig. 2). Therefore, in order to establish the permissible changes in the structural parameter vector  of the cylindrical weights, the inequality constraint  min 6  6  max (6) is
introduced.
4. Simulation and experimental results
The swinging up problem of the redesigned Pendubot is solved
via the concurrent redesign approach described in Section 2. This
approach simultaneously nds both the switching interval vector
Dt Dt1    DtnDt  2 RnDt and the structural parameters of the
weights  mp1 ; mp2 ; lcp1 ; lcp2 T 2 R4 that minimize JDt;  
 PnDt jDtj j P4 Nl subject to the motion equation of the
tf N
l1
j1
the
initial
position
Pendubot-Motor
x_ f xt; Dt;  ,
T
xo  p2 ; 0; 0; 0; 0 and the nal position xf p2 ; 0; 0; 0; 0T , assuming that there is a bangbang control ut dened in (7) where
umax 45 volts and umin 45 volts. The minimum and maximum
limit
vectors
 min mp1min ; mp2min ; lcp1min ; lcp2min T ,
 max mp1max ; mp2max ; lcp1max ; lcp2max T are explicitly dened as
and
 max 0:2; 0:135; 0:127;
 min 0; 0; 0:20; ;0:0381T

0:3048T . These limit vectors were established by a trial and error

procedure. This procedure consists of establishing an initial limit
vector  min ;  max . Then the DOP (8)(11) is solved by the projected
gradient algorithm (12) in order to nd the optimum structural
parameter  H . If an optimum structural parameter  H
i ; i 1; . . . ; 4
H


or



),
this
limit
value could
hits its limit value ( H
imin
imax
i
i
be enlarged if and only if this is physically possible. Then, other
optimization process which considers the modied limit vector
will be required for nding the optimum structural parameter vector. This procedure is repeated until the optimum structural
parameter vector  H does not hit the limit vector or even when
it is not physically possible to enlarge the limit vector.
The constants of the barrier functions x and the step sizes
kr ; kn required in (12), are selected as in (17), where diag is
the diagonal matrix with the elements  in the diagonal

x l 5E  8 for l 1 to 4

17

kr diag1E  3; . . . ; kr5;5 ; 3E  4; 3E  4; 5E  4; 5E  4
kn diag1E  4; . . . ; kn5;5 ; 1E  4; 1E  4; 1E  4; 1E  4
As the projected gradient algorithm could converge to a local minimum, several runs were performed in numerical simulations by
using different starting conditions  ini ; Dt ini , which were randomly
generated inside the feasible region imposed by  min and  max . The
starting conditions that generated the lowest value of the performance index (8) result as indicated in (18) and (19). With those initial starting conditions, the optimal redesign structural parameters
and the optimal switching intervals are shown in (20) and (21)

 ini 0:10; 0:034; 0:05; 0:1T

18

Dtini 0:08; 0:65; 0:4; 0:1; 0:05T

H

 0:1990; 0:0003; 0:1425; 0:1532

H

19
T

20
T

Dt 0:029; 0:614; 0:356; 0:088; 0:007

21

With the optimum parameters (20) and (21), the Pendubot brings
the links from the rest position to the upwards position in a miniP5
H
mum nal time of t H
j1 jDt j j 1:09 s. In (20), it is important
f
to note that the cylindrical weight of the link 2 should not be

Fig. 3. Pendubot with the optimal structural parameters.

182

C.A. Cruz-Villar et al. / Mechatronics 19 (2009) 178183

considered, as its mass mH

p2 is almost equal to zero. Then, the mass
mH
p1 and the distance between the rotation axis and the center of
H
mass lcp1 at the rst link are the structural parameters which affect
the positioning time of the Pendubot.
H
To implement the above results experimentally and since lcp1 is
negative, a rectangular support with small mass is added in order
to place the cylindrical weight of the rst link on the direction x
(Fig. 2). This support is made of aluminum with a mass of
0:035 Kg and it is shown in Fig. 3.
A modeling error of the Pendubot can produce nonoptimal design parameters for the real system, which could not appropriately
carry out the positioning task. Therefore, the optimization algorithm is computed again in a series of experimental trials, but
now with the state trajectory measured from the real system instead of the computed trajectory. Hence, once we have found the
optimal simulation results, the gradient projection algorithm (12)
must simultaneously compute both the optimal switching interval
vector Dtcnew and the optimal structural parameter  cnew for the
experimental results, but now with the real state trajectory and
the real nal state error. This procedure could be repeated if the
terminal state requirement ef  0 is not fullled. As a result, the

optimal parameters for the experimental setup are obtained. With

this procedure, the switching interval vector Dtcnew is the only that
signicatively changes. As opposite to the structural parameters
whose changes were innitesimally small, that is,  cnew   H which
were not physically feasible. The optimal switching interval vector
for the experimental setup is shown in (22), where Dt cH and  cH
represent the optimal design parameters for the experimental
results

 cH 0:1990; 0:0003; 0:1425; 0:1532T

DtcH 0:07; 0:58; 0:245; 0:06; 0:001T

The optimal switching interval vector and the optimal structural

parameters (22) for the experimental setup produced a swinging
P
5j1 jDtcH
up time tcH
f
j j 0:954 s.
The objective of the concurrent redesign approach (C. R. approach) is to adjust the structure-control parameters in order to
bring the system from one place to another in minimum time,
but when the system is in an environment with external forces,
specially the gravity force, the concurrent redesign approach does
not guarantee that the system permanently stays in the last position. So, a LQR control is used in order to stabilize the state vector

X1 (rad)

2
0
2
0

0.5

1.5

2.5

0.5

1.5

2.5

0.5

1.5

2.5

0.5

1.5

2.5

0.5

1.5

2.5

0.5

1.5

2.5

X2 (rad)

2
0

X3 (rad/s)

2
10
0

X4 (rad/s)

10

10
0
10

X5 (A)

5
0
5
50

u (V)

22

0
50

Fig. 4. Experimental results to swing up and stabilize the Pendubot with the concurrent redesign approach and a LQR control, respectively.

C.A. Cruz-Villar et al. / Mechatronics 19 (2009) 178183

Table 2
The simulation and experimental results of the concurrent redesign approach
C. R. approach tf
Simulation results
Experimental results without LQR
Experimental results with LQR

1:09 s
0:954 s
0:895 s

xt at the state 
x xtf p2 ; 0; 0; 0; 0T . The LQR gains K pm are obtained by using a linearized model of the Pendubot around
the state 
x and the lqrd Matlab function. Hence, K pm
403:5333, 421.7918, 91.8461, 64.0117, 2.9301] are the optimal LQR gains.
The switching of the control law from the one obtained by the C.
R. approach to the LQR, is carried out when the links are in the
attraction region of the LQR, given by, jex1 j 6 0:13963
rad; jex2 j 6 0:10472 rad; jex3 j 6 4 rad
 and jex4 j 6 8 rad
, where
s
s
exi xi t  xfi , which was experimentally obtained as the one that
minimizes the norm of the attraction region, given by the term
q
e2x1 e2x2 e2x3 e2x4 . Different attraction regions would lead to
different experimental positioning times. However, we are interested in the attraction region allowing the Pendubot to stay the
longer time being controlled by the bangbang control instead
the LQR control.
In Fig. 4, we show the experimental results of the C. R approach.
At the rst second, the open-loop bangbang control uDt with the
optimal switching interval vector (22) is applied. At time
t 0:895 s, the states of the Pendubot are inside the attraction region, then the LQR control replaces the bangbang control, stabilizing the links at 
x. Then the required time to switch the control law
is 0:895 s. Due to the attraction region of the LQR, the nal time in
this experiment is shorter than t cH
f .
In Table 2 the summary of the simulation and experimental
results of the concurrent redesign approach is shown. The nal
time of experimental results is enhanced because the optimal
design parameters Dt cH and  cH were modied when the gradient projection algorithm was computed again with the state
trajectory feedback from the real system, instead of the simulated trajectory. It should be pointed out that the experiment
was conducted several times. In order to give a statistical measure of the solution behavior, a fty sample set provided an
average nal time of t f 0:895 s and a standard deviation of
0:0204 s.
5. Conclusion
In this paper, a concurrent redesign approach that involves
structural and a bangbang control law parameters is proposed.
This approach reformulates the concurrent redesign problem as a
dynamic optimization problem, where the independent variables
are the structural (mechanical) and control parameters of an electromechanical system. To solve such a dynamic optimization problem, the projected gradient algorithm is used, therefore, all
functions involved in the redesign process, should be at least once
differentiable.
The concurrent redesign approach has been illustrated with a
highly nonlinear underactuated double pendulum system called
the Pendubot.

183

A weight of cylindrical shape is placed at each link in order to

change the dynamic parameters m1 ; m2 ; lc1 ; lc2 ; I1 ; I2 of the Pendubot. Then, the structural parameters to be modied are the masses
mp1 ; mp2 and the distances between the rotation axis and the center
of mass lcp1 ; lcp2 of the weights of cylindrical shape.
The concurrent redesign approach nds the switching interval
vector of the bangbang control law and the structural parameters
of the weights of cylindrical shape in order to minimize the required time to bring the redesign Pendubot from the rest position
xt0 x0 to the upward position xtf xf in minimum time.
If the system has already been built or not, with this concurrent
redesign approach it is possible to improve the performance of an
electromechanical system. When the system has already been
built, it is necessary to add a subsystem in order to modify the
structural parameters of the original system. When the system is
not constructed, it is necessary to select the shape (structure) of
the system and hence the structural parameters that the designer
wishes to optimize.
Although the concurrent redesign approach is illustrated with
an underactuated manipulator, it is considered that this approach
can be easily extended to other kinds of electromechanical systems. Specially, for redundant manipulators, new congurations
could be obtained by avoiding obstacles or state constraints.
Acknowledgments
The rst and second author acknowledge support from the
mexican Consejo Nacional de Ciencia y Tecnologa (CONACyT),
via the project number 084060. The third author acknowledges
support from CONACyT through a scholarship to pursue graduate
studies at CINVESTAV-IPNs Electrical Engineering Department.
References
[1] Sadegh N, Driessen B. Minimum time trajectory optimization and learning.
ASME J Dynamics Syst Measure Control 1999;121(2):2137.
[2] Bryson AE. Dynamic optimization. Upper Saddle River, NJ: Prentice-Hall; 1999.
[3] Binder T, Cruse A, Cruz-Villar CA, Marquardt W. Dynamic optimization using a
wavelet based adaptive control vector parameterization strategy. Comput
Chem Eng 2000;24:12017.
[4] Vakhrameev SA. A bangbang theorem with a nite number of switchings for
nonlinear smooth control systems. J Math Sci 1997;85(3):200216.
[5] Fu K, Mills JK. Convex integrated design (CID) method and its applications to
the design of a linear positioning system. IEEE Trans Control Syst Technol
2005;13(5):7017.
[6] Rastegar JS, Liu L, Yin D. Task-specic optimal simultaneous kinematic,
dynamic and control design of high-performance robotic systems. IEEE/
ASME Trans Mechatron 1999;4(4):38795.
[7] Ravichandran Th, Wang D, Heppler G. Simultaneous plant-controller design
optimization of a two-link planar manipulator. Mechatronics 2006;16(3
4):23342.
[8] Alvarez J, Cruz CA, Portilla EA. Evolutionary dynamic optimization of a
continuously variable transmission for mechanical efciency maximization.
Lect Notes Comput Sci 2005;(3789). ISSN 0302-9743.
[9] Fantoni I, Lozano R. Non-linear control for underactuated mechanical
systems. Springer-Verlag; 2002.
[10] Bazaraa MS, Sherali HD, Shetty CM. Nonlinear programming. New York: John
Wiley & Songs, Inc.; 1993.
[12] Gonzlez HG, Alvarez J, Alvarez-Gallegos J. Experimental analysis and control
of a chaotic pendubot. Int J Robot Research 2004;23(9):891901.
[13] Block JD. Mechanical design and control of the Pendubot. Master thesis,
General Engineering Department, Illinois University; 1991.
[14] Villarreal MG. Parametric redesign of Pendubot to vertical positioning in
minimum time [in spanish]. Master Thesis, Electrical Engineering Department,
CINVESTAV-IPN, Mxico, DF; 2005.
[15] Hibbeler RC. Engineering mechanics: statics & dinamics. New Jersey: Prentice
Hall; 1998.